Application of a Regional Screening Index for ... - ACS Publications

Chapter 11

Downloaded by NEWCASTLE UNIV on July 2, 2016 | http://pubs.acs.org Publication Date (Web): September 22, 2014 | doi: 10.1021/bk-2014-1168.ch011

Application of a Regional Screening Index for Chemical Leaching to Groundwater Vulnerability Analysis in the National Level S. J. Ki1 and C. Ray*,2 1Department of Civil and Environmental Engineering, University of Hawaii at Manoa, Honolulu, Hawaii 96822, United States 2Nebraska Water Center, University of Nebraska, Lincoln, Nebraska 68583, United States *E-mail: [email protected]

This chapter describes a regional screening index for use in Hawaii which has been recently updated to assess groundwater vulnerability to contamination by volatile organic compounds and pesticides. Specifically, this chapter will discuss two issues: 1) how we assess the accuracy of a new screening index compared to an analytical solution for the movement of contaminants and 2) how we extend the regional index scheme into a national-scale vulnerability assessment. We found that the screening index was able to be used as an initial diagnostic tool for groundwater vulnerability assessment as it consistently provided a conservative estimate of the vulnerability to protect public health. The results of the national-scale assessment also showed that the groundwater vulnerability to agricultural chemicals varied widely among the conterminous 48 states, which demonstrated the feasibility of the regional screening index to a large-scale vulnerability assessment. We present two examples to illustrate the regional and national levels of groundwater vulnerability assessment.

© 2014 American Chemical Society Jones et al.; Describing the Behavior and Effects of Pesticides in Urban and Agricultural Settings ACS Symposium Series; American Chemical Society: Washington, DC, 2014.


Introduction Agricultural chemicals such as pesticides and volatile organic compounds (VOC) are the major contaminants that impact national water quality in the United States (US) (1, 2). Pesticides applied to or present near the soil surface undergo two primary processes: moving with runoff and leaching down the soil profile, respectively, and they have wide impacts on surface water and groundwater quality (1, 3). During subsurface transport, pollutants move through soils with percolating water and eventually reach groundwater if they are not sufficiently attenuated by sorption and degradation within given soil layers (4, 5). Chemical properties of a pollutant as well as environmental conditions such as soil and recharge characteristics are typically found to be the major factors that determine the degree of attenuation in soil (6). Various index approaches, in contrast with the DRASTIC method that assesses aquifer vulnerability without consideration for individual pollutant characteristics, have been suggested to estimate the amount of attenuation of a chemical in subsurface transport (7–9). This index strategy is particularly useful when model parameters required to provide a detailed description of subsurface processes are difficult to obtain or are not available. Attenuation factor (AF) is one of popular indices, listed in initial pesticide evaluation step in Hawaii, that estimates the leaching potential of pollutants using three important properties, such as sorption, degradation, and recharge as discussed above (4–7). Other accepted methods are Screening Concentration In GROund Water (SCI-GROW) and Windows Pesticide Screening Tool (WIN-PST) which are recommended by US Environmental Protection Agency (US EPA ,) and US Department of Agriculture (USDA ,), respectively. Like the AF, these two indices use similar properties to examine aquifer vulnerability to pesticide contamination through either regression analysis or rating scale appraoch. Some of the main drawbacks with these two methods are a lack of their generalization ability and reliance on non-volatile compounds. Recently, the State of Hawaii has adopted a regional screening index that revises the original AF for identifying vulnerable areas to VOC plus pesticides. This was done to account for volatilization loss which reduced the amount of pollutant leaching through soils as it was not addressed in the previous index approaches (10). Using the regional screening index, this chapter will explain: 1) how the contamination risk of volatile and non-volatile chemicals is evaluated at the state level, 2) how much accuracy can be expected from the screening index under local conditions in Hawaii, as compared to an analytical solution of volatile chemical transport, and 3) what steps should be taken to address the national level of groundwater vulnerability. With the examples presented, this chapter provides insight into the basic process of soil contamination and the application of geographic information system (GIS) tool in vulnerability analysis.

276 Jones et al.; Describing the Behavior and Effects of Pesticides in Urban and Agricultural Settings ACS Symposium Series; American Chemical Society: Washington, DC, 2014.

Materials and Methods Regional Screening Index


The regional screening index (RSI) is a modification of the mass fraction model (i.e., Mr/M0) that has been initially developed for evaluating the leaching of volatile chemicals in a dual-porosity model which divides soil porosity into two domains, macro- and micro-pores (10). After assuming 1 m of effective root zone as well as neglecting the terms water uptake (by root) and diffusive loss (to micro-pore) from the original equation of Hantush et al. (10), we can arrive at the following equation:

where, M0 and Mr are initial mass at the soil surface and residual mass at a reference depth d (m), respectively. The chemical-related parameters T1/2, Koc, Kh, and Dg, respectively, indicate the half-life (d), sorption coefficient (m3/kg), dimensionless Henry’s law constant (–), and gaseous diffusion coefficient (m2/d). The soil-related properties ρb, foc, θFC, and na signify the bulk density (kg/m3), organic matter fraction (–), moisture content at field capacity (–), and air-filled porosity (–), respectively. Lastly, q and l represent the groundwater recharge rate (m/d) and boundary layer thickness on top of the soil surface (m), respectively.

Compiling Databases As shown in eq 1, the parameter information is required to evaluate the contamination risk of chemicals to groundwater. First, chemical properties of test compounds were obtained from around 800 references such as national and international pesticide databases (11–13). Table I shows a summary of key characteristics of 10 example compounds used for the groundwater vulnerability assessment. Next, two different levels of digital soil data (i.e., detailed and general soil maps in the US) which were retrieved from an online database were used for the groundwater vulnerability assessment at the state and national levels, respectively (14). This is because application of the detailed soil data to the 48 lower states needs a large amount of data storage capacity. Then, groundwater recharge maps for the State of Hawaii and the US were collected from state (15) and federal agencies (16), respectively. Specifically, the national recharge map was created to estimate contaminant loads (e.g., suspended sediments and nutrients) in US streams, whereas the statewide recharge map was generated from a numerical simulation study carried out under Hawaii’s source water assessment program. Updating all the parameters enabled assessment of the vulnerability to groundwater contamination, which was then compared with the result of the 277 Jones et al.; Describing the Behavior and Effects of Pesticides in Urban and Agricultural Settings ACS Symposium Series; American Chemical Society: Washington, DC, 2014.

STANMOD program. The STANMOD contains a series of analytical solutions derived from advection and dispersion equations in soils including degradation and volatilization (17). Three required parameters Koc, T1/2, and Kh plus the value of Dg for each compound were identically used for the STANMOD, as applied to the regional screening index (see Table I). In both cases, l and d were set to 0.05 m and 0.5 m, respectively.

Results and Discussion


Regional-Scale Groundwater Vulnerability Assessment A regional screening index (see eq 1) was applied to assess the groundwater vulnerability in the State of Hawaii. Figures 1a and 1b illustrate the contamination risk of two islands in Hawaii: chlorobenzene on the island of Molokai and 2,4-D on the island of Hawaii, respectively. The chemical properties of the two compounds used for the state-wide groundwater vulnerability assessment are described in Table I. These figures show that the chemicals are classified into four types of contamination levels: 1) no data (if information on the soil bulk density or moisture content is not available), 2) unlikely, 3) uncertain, and 4) likely. This risk categorization is established by a classification method which compares the contamination risk of a particular compound to that of a known leacher (e.g., atrazine) and a non-leacher (e.g., endosulfan determined from local groundwater monitoring studies. From the figure, it was determined that while the contamination risk of test chemicals in the two islands appeared to be generally uncertain, a high and low contamination risk was also observed sporadically in some areas. This is because local conditions such as the recharge and soil characteristics vary across different areas even though the chemical properties of the test compound are identically maintained in each island. Therefore, it can be suggested from these examples that the regional screening index can be successfully used for the regional-scale groundwater vulnerability mapping as long as it provides a reasonable estimate of pollutant leaching versus other similar models. Note that a careful review of reference chemicals (i.e., a leacher and non-leacher) from the monitoring studies is needed to increase its predictive validity at the state level as the index itself is not used as an independent prediction of leaching.



Model Validation with STANMOD The STANMOD program, which provided a closed-form solution for predicting contaminant movement in soils, was used to examine accuracy of the regional screening index. Three soils on the island of Oahu were selected for validation of the regional screening index (see Figure 2a) because these soils were located inside the capture zones that were considered important to protect public water supply wells and groundwater resources (15). Here, the capture zone indicates the virtually delineated boundary within which groundwater in specific areas can travel to the wells. Table II presents the physical properties of three soils A, B, and C on this island which are averaged over topsoil, from the soil surface to a depth of 0.5 m. The recharge rates in the soils were obtained from the groundwater modeling study that was conducted to simulate flow and pollutant transport processes including the capture zone delineation, as discussed above (15). Figure 2b presents a result of comparison between the STANMOD and the regional screening index. In this figure, each data point indicates the contamination risk of 10 individual chemicals on three test soils (see Table I). It was found that the STANMOD typically overestimated the contamination risk of compounds than the regional screening index. It can be assumed that the STANMOD is generally more accurate than the regional screening index between low and medium recharge conditions, although both are not expected to outperform complex models. This is becuase the regional screening index is derived under the condition of an infinite Peclet number, i.e., advection dominated flows (10). The difference between them was smaller in Soils A and C than Soil B, as explained by the coefficient of determination, R2. From the result, it is concluded that the regional screening index can still be used to evaluate the contamination risk of chemicals at initial vulnerability screening level, because it provides at least a vulnerability level that is safe to protect public health in both best- and worst-case scenarios.



Figure 1. Examples of groundwater vulnerability assessment using the regional screening index: (a) chlorobenzene in Molokai Island and (b) 2,4-D in Hawaii Island.


Table I. Chemical properties of 10 compounds to compare the performance of the regional screening index and the STANMOD Compoundsa

KOC (m3/kg)

T1/2 (days)

Kh (–)

Atrazine*, b

0.126

61.6

2.50 × 10-7

0.472

43.4

6.29 ×

10-4

0.342

1.23 ×

10+0

0.654

1.53 ×

10-1

0.628

4.28 ×

10-2

0.846

1.01 ×

10-1

0.870

7.39 ×

10-1

0.560

7.78 ×

10-1

0.661

10-7

0.479 0.646

Carbofuran Carbon tetrachloride Chlorobenzene*


1,2-Dichloroethane DCM PCE 1,1,1-Trichloroethane

0.048 0.146

148.8

0.260

73.4

0.044

86.0

0.018

23.8

0.279

270.0

0.248

259.3

Dg (m2/day)

2,4-D*

0.053

11.7

6.22 ×

Xylenes (total)c

0.376

360.0

2.12 × 10-1

a Other chemical names: DCM = Dichloromethane, PCE = Tetrachloroethylene, and 2,4-D = 2,4-Dichlorophenoxyacetic acid. b The chemicals used for regional and national vulnerability assessment are indicated by asterisk (∗). c Chemical characteristics of xylenes (total) were compiled from two isomers of p- and m-xylenes.

Figure 2. Validation of the regional screening index in assessing the contamination risk of chemicals: (a) three test soils in Oahu Island and (b) a comparison between the regional screening index and the STANMOD. Individual symbols in Figure 2b indicate different chemical compounds examined for three test soils (see Table I).


Table II. Physical properties of three test soils at 0.5 m depth used for validation of the regional screening index (see Figure 2a) MUKEYa

N (–)

ρb (kg/m3)

θFC (–)

fOC (–)

qb (m/day)

Soil A

468510

0.534

1,330

0.391

0.026

2.33 × 10-3

Soil B

468393

0.615

1,100

0.303

0.021

2.05 × 10-3

Soil C

468533

0.667

950

0.173

0.121

6.72 × 10-3

Soils

MUKEY refers to the map unit key (i.e., the soil indentifier) that associates soil polygons with tabular data, the physical and chemical characteristics of soils. b The recharge rates for specific soils are obtained from a Source Water Assessment Program in the State of Hawaii (15).


a

National-Scale Groundwater Vulnerability Assessment Finally, the regional screening index was extended to assess groundwater vulnerability outside Hawaii as it only required the minimum number of parameters easily available (see eq 1). Figure 3 illustrates the procedure for evaluating the groundwater vulnerability to atrazine in the State of California in the US, where the soil and recharge information, except for the same chemical properties, is newly added to the regional screening index. In this example, the contamination risk of a chemical is classified into six types of contamination levels depending on the value of the regional screening index: from no data (0) through very low (< 0.0001) to very high (> 0.25). This is because selecting reference chemicals simply from the national groundwater monitoring studies may not correctly reflect the results of statewide monitoring data. In fact, some differences were observed in the contamination risk of atrazine in the State of California between the regional screening index and the regional (18) and national monitoring studies (1). The main reason is that the regional screening index does not account for heterogeneous geoenvironmental conditions, history of pesticide use, and groundwater level in the state and national levels. In addition, the regional screening index does not assess groundwater quality in monitoring wells. As discussed in the regional vulnerability assessment, a more rigorous procedure for selecting reference chemicals is, therefore, needed to ensure its prediction accuracy at the national level. However, implementation of such complex models within the GIS framework is not easy and we leave this for future study. In this way, the new databases containing soil and recharge properties in the 48 contiguous states were created for the national-scale groundwater vulnerability assessment. Figure 4 presents the contamination risk of atrazine in the lower 48 US states (see a solid black bar). In the figure, a gray bar indicates the number of soil polygons in each state compiled from a general soil map in the US. Based on the number of soil samples in the states, the mean and (95%) confidence interval of the regional screening index were estimated. It was shown that there existed a high degree of variation in the contamination risk of atrazine in individual states. Among them, three states such as California, Washington, and Montana represented the highest vulnerability to atrazine due largely to a high 282 Jones et al.; Describing the Behavior and Effects of Pesticides in Urban and Agricultural Settings ACS Symposium Series; American Chemical Society: Washington, DC, 2014.


groundwater recharge provided (16). In Arizona, soil properties appeared to be more vulnerable to leaching than the remaining states. Conversely, some states such as Indiana and Minnesota also showed a high pollutant attenuation capacity resulted from the combined effect of chemical, recharge and soil properties. However, as the national groundwater monitoring studies showed medium to high occurence of this compound in those regions (1), additional investigation on the leaching mechanism is required. From this result, it is confirmed that the regional screening index can be used to the national-scale groundwater vulnerability assessment, although some efforts to adjust its prediction with the monitoring data appear to be needed. Finally, this will allow relative risk of contaminants between states to be rapidly assessed and summarized in GIS map.

Figure 3. A procedure for assessing groundwater vulnerability on the national level. Shown in the sample is the contamination risk of atrazine in the State of California in the United States.



Figure 4. The contamination risk of atrazine at the lower 48 states in the United States (see the left/bottom axis). An error bar indicates the 95% confidence interval. A bar graph shown as gray presents the number of soil polygons in individual states (see the top/right axis).

Conclusion In this study, the regional screening index that evaluates the leaching of volatile and non-volatile chemicals is applied to the state- and national-scale groundwater vulnerability assessment, i.e., in Hawaii and the conterminous 48 states. A few simple parameters were used to describe the contamination risk of agricultural chemicals to groundwater. Appropriate databases at the state and national levels were constructed to reflect variation in environmental conditions. Below are the major findings of the study. 1.

2.

3.

The regional screening index was successfully applied to the state-wide groundwater vulnerability assessment in Hawaii. Groundwater vulnerability varies between the chemicals and local conditions such as the recharge and soil characteristics. The accuracy of the regional screening index was tested with an analytical solution of the STANMOD program. Although the regional screening index shows weak agreement with the analytical solution, it is likely to offer an aquifer vulnerability in the safe level to the public by providing a conservative leaching potential than other models in the best condition for leaching. Compiling new databases such as the soil and recharge information at the national level enabled the regional screening index to evaluate the contamination risk of chemicals in the 48 contiguous states. As shown in the statewide vulnerability assessment, the contamination risk differs considerably among states and each chemical. Therefore, we suggest a 284

Jones et al.; Describing the Behavior and Effects of Pesticides in Urban and Agricultural Settings ACS Symposium Series; American Chemical Society: Washington, DC, 2014.

new method be developed to evaluate the groundwater vulnerability to agrochemical contamination at both state and national levels, specifically for those that show dissipation and/or volatile emission in soils.

References


1.

2.

3.

4. 5. 6. 7. 8. 9. 10. 11. 12.

13. 14. 15. 16.

Gilliom, R. J.; Barbash, J. E.; Crawford, C. G.; Hamilton, P. A.; Martin, J. D.; Nakagaki, N.; Nowell, L. H.; Scott, J. C.; Stackelberg, P. E.; Thelin, G. P.; Wolock, D. M. The Quality of Our Nation’s Waters–Pesticides in the Nation’s Streams and Ground Water, 1992–2001; USGS Circular 1291; U.S. Geological Survey: Reston, VA, 2006. Zogorski, J. S.; Carter, J. M.; Ivahnenko, T.; Lapham, W. W.; Moran, M. J.; Rowe, B. L.; Squillace, P. J.; Toccalino, P. L. The Quality of Our Nation’s Waters–Volatile Organic Compounds in the Nation’s Ground Water and Drinking-Water Supply Wells; USGS Circular 1292; U.S. Geological Survey: Reston, VA, 2006. State Source Water Assessment and Protection Programs Guidance, Final Guidance; EPA 816-R-97-009; Office of Water, U.S. Environmental Protection Agency: Washington, DC, 1997. Loague, K.; Bernknopf, R. L.; Green, R. E.; Giambelluca, T. W. J. Environ. Qual. 1996, 25, 475–490. Alavi, G.; Dusek, J.; Vogel, T.; Green, R. E.; Ray, C. Vadose Zone J. 2007, 6, 735–745. Stenemo, F.; Ray, C.; Yost, R.; Matsuda, S Pest Manage. Sci. 2007, 63, 404–411. Diaz-Diaz, R.; Loague, K. J. Am. Water Resour. Assoc. 2000, 36, 823–832. Cohen, S. Groundwater Monit. Rem. 1998, 18, 55–57. Brown, J.; Bach, L.; Aldous, A.; Wyers, A.; DeGagne, J. Front. Ecol. Environ. 2011, 9, 97–102. Hantush, M. M.; Govindaraju, R. S.; Marino, M. A.; Zhang, Z. L. J. Hydrol. 2002, 260, 58–74. Wauchope, R. D.; Buttler, T. M.; Hornsby, A. G.; Augustijnbeckers, P. W. M.; Burt, J. P. Rev. Environ. Contam. Toxicol. 1992, 123, 1–155. Soil Screening Guidance: Technical Background Document; EPA 540-R95-128; Office of Emergency and Remedial Response, U.S. Environmental Protection Agency: Washington, DC, 1996. PPDB: Pesticide Properties DataBase. University of Hertfordshire. http:// sitem.herts.ac.uk/aeru/ppdb/en/index.htm (accessed October 17, 2013). Web Soil Survey. U.S. Department of Agriculture. http:// websoilsurvey.nrcs.usda.gov/ (accessed September 30, 2013). Whittier, R. B.; Rotzoll, K.; Dhal, S.; El-Kadi, A. I.; Ray, C.; Chang, D. Hydrogeol. J. 2010, 18, 711–723. Gilliom, R. J.; Alley, W. M.; Gurtz M. E. Design of the National WaterQuality Monitoring Program–Occurrence and Distribution of Water-Quality Conditions; USGS Circular 1112; U.S. Geological Survey: Sacramento, CA, 1995. 285

Jones et al.; Describing the Behavior and Effects of Pesticides in Urban and Agricultural Settings ACS Symposium Series; American Chemical Society: Washington, DC, 2014.


17. Jury, W. A.; Spencer, W. F.; Farmer, W. J. J. Environ. Qual. 1983, 12, 558–564. 18. Burow, K. R.; Stork, S. V.; Dubrovsky, N. M. Nitrate and Pesticides in Ground Water in the Eastern San Joaquin Valley, California : Occurrence and Trends; USGS Water-Resources Investigations Report 98-4040; U.S. Geological Survey: Sacramento, CA, 1998.


Application of a Regional Screening Index for ... - ACS Publications

Recommend Documents